Node with two wireless interfaces - to bridge them or not?

I have one node that has two wireless interfaces. I was thinking to increase somewhat the multi-hop bandwidth loss and what is the best way of doing that. Currently I have ath0 with two nodes on there and ath1 with another two nodes but I'm not bridging those two interfaces. The thing is, they are on the same subnet so is there any point in keeping them "separate", any performance advantage, or should I just make a bridge and add the two ath's to it and just have a single IP and run batman on that one bridge or leave it as it is, with two IPs from the same subnet and run batman ath0 /w ath1 /w as I do currently?

Pele

Hello Pele,

bridging multiple interfaces is probably not a good idea, because as soon as you have a cycle there will be bridging loops. The frames will cycle until they are droppend and there is no TTL to stop them! :) In Ethernet, a solution for this is STP [1], which will cut off (one of) the links in the loop, but probably the good ones, so that is not what you want. That is what the mesh is supposed to do. But bridging would not bring you any performance increase anyway, so i'd suggest to stick with your current setup.
(if you want to have only IP per interface for some reason, you can also have a look at batman-advanced).
To increase multi-hop bandwidth you should better try the standard wifi ways to reduce interferences/packet loss. E.g. try to switch off rts, switch on fragmentation, limit baserate etc.

Best regards
Simon

[1] http://de.wikipedia.org/wiki/Spanning_Tree_Protocol

Thanks, so I'll keep it as it is then as far as interfaces are concerned.

I would love to try out batman-adv with layer2 (as I truly believe that is the only way forward in meshes) but as it stands all my routers are on rooftops and if I mess something up I'm stuck! This way, I just keep updating batmand and it works (but not as fast as I'd like it to be).

I've tried to play with FRAGmentation but setting it to anything other than off (2346) the transfer (I use wget ftp://host/file) stops.
I played with RTS and I really don't know what to say. It is SO FRUSTRATING! Yesterday I've had the mesh set up and it ran OK (1.7Mbps between two nodes - is there ANYONE that gets 11Mbps, EVER?) but today nothing works. And it just keeps driving me crazy. Every day I have to reset and re-test and fiddle with it. I just need settings that work. I just can't believe that with my rssi of 17-20, which looks decent enough, I cannot get consistent performance. I believe madwifi drivers also have a big role to play in my frustration. Also this mixture of Broadcom and atheros is driving me crazy. 4 atheros and 1 broadcom, and Broadcom just keeps quitting. After a day or two it disassociates and disappears, then I have to go over and unplug it. MAD!

I wish there was someone out there who knows wireless inside-out and who would be able to spare a day in helping me with getting this thing going. As it stands, it just makes me want to give it all up and become a farmer or something.

Pele

Hi -

Fragmentation is broken in current Madwifi. Every packet bigger than the fragmentation value will not be send. This is bad, because fragmentation is pretty important for smooth data transfers.

Another notorious problem is IBSS cell-splitting, you can fix that by using

iwconfig athX ap 02:CA:FF:EE:BA:BE (or whatever cell-id you wish)

Also it is recommended to select the ministrel rate adaption mechanism when loading the driver.

Try a recent Madwifi version from OpenWRT Kamikaze (i.e. get the OpenWRT patches and apply them to the Madwifi version used in OpenWRT).

cu elektra

Hi,

my name is Solon and I'm trying to build a wifi network for the Studentendorf Schlachtensee in Berlin. For various reasons, I'd like to integrate BATMAN Advanced into the ROBIN-FW or any other FW that makes deployment
through people without any IT-knowledge possible.

My question is: are there any developers or people with profound knowledge, who attend the c-base meetings on wednesdays and who are willing to give me an inside look into the recent builds?

Thx in advance...

solon

Hi -

I usually spend my wednesday evening there.

cu elektra

Hi,

I think the "batman advanced" topic is quite new and you wont find people that can help you easily. The first weekend in May we will have a wireless community weekend in Berlin (C-Base). I will be there and so will be Simon. We are the main developers behind the layer 2 implementation.
I believe more people are interested in that topic - may be we will do a little workshop ?

Greetings,
Marek

Dear Marek Lindner,

A BATMAN adv workshop would be great, if we need a little bit more space, 30+ fonera APs and a huge area - I can provide all of it...

bye

solon

Hi,

sounds very cool - we should do that.

@cven: Can you help preparing that ?

Greetings,
Marek

ML> Hi,

More to come:

Have a look at www.studentendorf.de , if we meet there, we can use all the existing facilities:

Internet-Access, a huge meeting-room with enough tables and chairs for 50+ people, Student-Club www.club-a18.de (food/drinks from 20:00 to 1:00).

Free accommodation for people with mattresses/sleeping-bags can be arranged, breakfast for everybody is also possible. You can also rent a bed&breakfast room on site for €24.

Public transportation:

BUS 112/N48 or S-Bahn Schlachtensee (7min) / Nikolassee (12min)

bye

Solon

Hi -

at the Wireless Community Weekend (May 2nd-5th in Berlin/Germany @ C-Base) there will be a test with 20+ Foneras running RO.B.I.N.

The B.A.T.M.A.N. developers will be around (maybe we'll miss Axel, he wasn't sure whether he will be in Berlin on time).

If you have the time, join us!

cu elektra

WCW Venue:
c-base e.V.
rungestrasse 20
10179 berlin (center)

Hi -

correction: May 2nd-4th not May 2nd-5th.

cu elektra

Hello Solon,

thank you for the offer, but i really think we should do this workshop in the c-base, because all people will be assembled there, and i'm sure there are some people who will join us spontanously.

It would be nice if you could bring some FON-routers (like 5 to 10) that we can use for the workshop for the batman-adv. That would be enough for the start, and maybe more people will come and join with their routers.

Best Regards,
Simon

SW> Hello Solon,

SW> It would be nice if you could bring some FON-routers (like 5 to 10) that SW> we can use for the workshop for the batman-adv. That would be enough for SW> the start, and maybe more people will come and join with their routers.

I'll bring plenty of FONs with me, but I can't be there on friday. Expect my arrival on saturday 11-12am.

Regards, Solon

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

stupid question:
will this be a batman l2 test or a robin test?

Alex
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFIGgt3hx2RbV7T5aERAqf6AJ959H9a5OQSwDwXB1UtQ8RydwI/zwCfaVx5 SiruhziEi1k/OgqLYEyetsw=
=pWTd
-----END PGP SIGNATURE-----

A. Try out BATMAN-L2
B. Try to integrate it into the ROBIN-FW

Best regards,

Solon Lutz

+-----------------------------------------------+
| PyroDesign Berlin - Creativity for tomorrow   |
| Wasgenstrasse 75/13 - 14129 Berlin, Germany   |
| www.pyro.de - phone + 49 - 30 - 48 48 58 58   |
| [email protected] - fax + 49 - 30 - 80 94 03 52    |
+-----------------------------------------------+

hi

will it be some day technically possible to use the marvel wifi chip of the laptop to join the mesh network or is always a router needed?

Why can´t that drawn to the laptops with a software solution ?

Max

Hi Max -

this is solely a issue of malfunctioning drivers/ WiFi cards in ad-hoc mode. The proprietary Broadcom driver for 2.4 Linux kernels shipped with devices like Linksys WRT54GL, Buffalo WHR-G54S, Asus WL500GP works flawlessly.

The Madwifi driver for Atheros cards works but still has issues that developers are working on. The best working Madwifi driver seems to come with OpenWRT.

This is basically what works at the moment. Most Intel drivers have issues (there are firmware issues as well, as I have experienced with IPW2200). IPW2100 seems to work, IPW3945 works more or less with some tricks (and occasional firmware errors).

Marvel drivers are nowadays shipped with the Linux kernel, but because I lack Marvell hardware I couldn't test it in ad-hoc mode.

In general the ad-hoc mode has been widely neglected by manufacturers and developers. From the first day I started to work on mesh networks I have been battling with firmware/driver issues. But as more and more people start using mesh networks the demand for working drivers is increasing. So we can expect that - finally, after many years - the situation will improve quickly.

cu elektra

Hi Phillipe -

well, Freifunk and others around the globe are successfully using mesh as network infrastructure for their daily communication needs. (I'm using multihop mesh as internet uplink since 4 years).

Here is how it works: Rather than using a zoo of different hardware use one that works, and that is the proprietary (I hate to support this!!!) Broadcom driver for the embedded Broadcom designs used in many 802.11b/g routers.

Get this and go ahead deploying your network. You'll love it. Particularly if you use Freifunk firmware. This is one key to success nowadays.

It is a pitty that Linksys USA is not selling WRT54GL's in the US (correct me if I'm wrong). So most people there are using Atheros SoC - Fonera, Meraki - which are all made by Accton/Taiwan. (Again, correct me if I'm wrong). While people in Europe, Africa and Asia are predominantly using WRT54GL.

As far as I understood Antonio Anselmi, recent RO.B.I.N. firmware works stable in ad-hoc with Atheros SoC.

I was working on a system for the Meraka institute in SA, for their 'Massive Mesh' grid, which was build to practically develop/test/verify mesh routing protocols and drivers. (I was running tests of OLSRD and all B.A.T.M.A.N. versions there.)

I have patched an older Madwifi which works rock-solid in the Meraka mesh. There is only one issue: Fragmentation does not work. To be
precise: If you enable it, the cards won't send data packets greater than the fragmentation value anymore :-(

AFAIK this issue is still not solved.

Running different Madwifi VAPs in ad-hoc and master mode is something I didn't test. Last time that I have tested it (quite a while ago) it was a hack. And I was happy if my machine didn't lock up after a while. Since I performed those tests several months ago I can't say much about it.

I see it that way: Keep infrastructure clients separated from your mesh infrastructure - particularly keep clients away from your mesh channel. Otherwise you will degrade performance by using and re-using the same channel all the time. Even if Madwifi VAPs work stable-ish this doesn't perform well. A stupid but working access point is cheap, use the mesh as an uplink. Well, if all that you want is a very cheap mesh with a couple of nodes and want to accept performance issues, you may be fine with VAPs if the driver works.

The new ath5k driver in the most recent Linux kernel works in ad-hoc, but has range issues. As could be read recently in the news, Atheros is actively sponsoring development of their hardware for the kernel now (they are paying a guy to develop drivers).

cu elektra

Uh, hum... I think this was cleared up, but I have a lot of WRT54gls (in Texas) and I was also able to buy the T-Mobile version with 32MB RAM.

Cheers,

D Davis

thanks, elektra, would it be possible, that an usb stick with wifi chip and coded in software can be distributed? e.g. by foebud?
we need a solution, where people use their laptop and not need to configure their routers.
The Laptop or USB-Stick has the wifi chip...
is there a wiki, where all wifi chips are listed and how the bug fixing for them is going on by whoom?

Maxx

Hi -

I can't recommend using a variety of different wireless hardware with different chipsets, drivers, firmware in a network - yet. And I don't know any USB stick that I would recommend. If someone reading this list has a working recommendation still available for purchase, please add a comment.

What we usually do is: Get a router supported by Freifunk/OpenWRT with Broadcom/Atheros SoC inside, reflash it with Freifunk Firmware and connect it to your notebook. Or: If you have Atheros in your notebook or PC with a self compiled recent Madwifi, fine. Otherwise replace the build-in card with something Atheros-based (you may have to modify your notebooks BIOS if you use IBM/Lenovo/HP, in order that the machine accepts the card), or insert a additional cardbus/express card with Atheros chipset, given that you still have a cardbus/express interface in your notebook. The Madwifi driver shipped with your favorite Linux distro may not work properly, so get it from madwifi.org and compile it against the Linux kernel that you are running.

With Atheros you can fix the IBSSID in order to overcome those notorious issues with cell-splitting, I also managed to perform this with IPW3945. Also OpenWRT/Freifunk does allow this with Broadcom hardware, too.

The command on a Linux-PC to fix the IBSSID is:

iwconfig <interface> ap <your preferred cell-id like 02:CA:FF:EE:BA:BE>

Note that this is non-standard behavior, and is unlikely to work with other cards. Actually upon request from Freifunk this was modified in the Madwifi driver.

Old Cisco Aironet 802.11b works for me (mostly) as long as the mesh is the only ad-hoc network around... I use it every day, it is running stable until I stop it. Also ancient Atmel 76c503 based USB sticks work - I used to build mesh routers with them in the old days ;-) Old Prism2, 2.5, 3 and Prism54 (Hard-Mac only) work, but are susceptible to cell-splitting (and may lock up if cell-splitting is already going on in your WiFi cell).

I have a Intel IPW3945 in a new notebook that works somehow with a little trick under Ubuntu (fix the Cell-ID in managed mode first, switch to ad-hoc, set essid, set channel, enable interface) - but from time to time I get hiccups that seem to be associated with firmware errors (reported in syslog) - then I need to reset the interface). The experience with IPW2200 was even worse with a notebook I used a year ago, but I didn't try that recently.

Broadcom would be an option, but the company doesn't provide drivers for Linux-PCs. There is an open-source driver but I don't know whether it actually works in ad-hoc mode now.

cu elektra

Hi Philippe -

That's good to know.

It is small - ok. Things I like about WRT54GL: JTAG (even if you hose the system completely including bootloader you can restore it), serial interface, wide range DC-input (5-20 Volts), doesn't die if you reverse polarity, low power consumption (3.6 Watts under full load). Up to 250mW transmit power. Makes it ideal for my solar-powered home and developing countries, outdoor nodes, disaster areas and so on. Open up the housing and have a look at those *three* individual DC-DC converters for every internal voltage.

Disclaimer: I am not working or getting paid by Linksys. I'm willing to utilize everything good that comes the way...

Of course I don't like using proprietary code, this is sad. You know Linksys/Cisco had to be forced to their own good. They must be selling those devices like crazy. We have WRTs running on churches in a noisy area with many competing networks, each one linked to 15-20 other single hop mesh nodes. Somewhere around 30 single hop neighbors there is a limit for the driver/chipset, I suppose. The driver works rock solid. As soon as Madwifi or ath5k (with OpenHAL) can compete with that I'll be much more than happy.

If I could utter a wish to a fairy coming my way I would also ask to have at least one good performing usb chipset for 802.11abgn that works with Linux as well. I'd start assembling omnis and directionals with usb-connector immediately ;-)

Cheers,

cu elektra

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

i would like to point out that felix from openwrt is looking for bugs to fix, as he is running out of madwifibugs.

so, if there is a openwrt-madwifi breakage, please contact him. <removed>

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFID2e1hx2RbV7T5aERAs0KAJ9vDu8GYiMAypqFh5B53YsX8lOjTACgspek Atwbw+u6sF+XBItWp65lgo0=
=yU16
-----END PGP SIGNATURE-----